ParameterStore Class Reference

Interface for object which can store parameters. More...

Detailed Description

Interface for object which can store parameters.

#include <ParameterStore.h>

Inheritance diagram for ParameterStore:

Public Member Functions
	ParameterStore (Model *object)
	ParameterStore (const ParameterStore &)=delete
	ParameterStore (ParameterStore &&)=delete
ParameterStore &	operator= (const ParameterStore &)=delete
ParameterStore &	operator= (ParameterStore &&)=delete
virtual	~ParameterStore ()=default
const std::map< std::string, std::unique_ptr< TensorValueBase > > &	named_parameters () const
std::map< std::string, std::unique_ptr< TensorValueBase > > &	named_parameters ()
const TensorValueBase &	get_parameter (const std::string &name) const
	}@
TensorValueBase &	get_parameter (const std::string &name)
	Get a writable reference of a parameter.
void	set_parameter (const std::string &, const Tensor &)
	Set the value for a parameter.
void	set_parameters (const std::map< std::string, Tensor > &)
	Set values for parameters.
virtual void	send_parameters_to (const TensorOptions &options)
	Send parameters to options.
template<typename T , typename = typename std::enable_if_t<std::is_base_of_v<TensorBase<T>, T>>>
const T &	declare_parameter (const std::string &name, const T &rawval)
	Declare a parameter.
template<typename T , typename = typename std::enable_if_t<std::is_base_of_v<TensorBase<T>, T>>>
const T &	declare_parameter (const std::string &name, const TensorName< T > &tensorname, bool allow_nonlinear)
	Declare a parameter.
template<typename T , typename = typename std::enable_if_t<std::is_base_of_v<TensorBase<T>, T>>>
const T &	declare_parameter (const std::string &name, const std::string &input_option_name, bool allow_nonlinear=false)
	Declare a parameter.
void	assign_parameter_stack (jit::Stack &stack)
	Assign stack to parameters.
jit::Stack	collect_parameter_stack () const
	Collect stack from parameters.

Constructor & Destructor Documentation

ParameterStore() [1/3]

ParameterStore

(

Model *

object

)

ParameterStore() [2/3]

ParameterStore ( const ParameterStore & )

delete

ParameterStore() [3/3]

ParameterStore ( ParameterStore && )

delete

~ParameterStore()

virtual ~ParameterStore ( )

virtualdefault

Member Function Documentation

assign_parameter_stack()

void assign_parameter_stack ( jit::Stack & stack )

protected

Assign stack to parameters.

collect_parameter_stack()

jit::Stack collect_parameter_stack ( ) const

protected

Collect stack from parameters.

declare_parameter() [1/3]

template<typename T , typename >

const T & declare_parameter ( const std::string & name, const std::string & input_option_name, bool allow_nonlinear = false )

protected

Declare a parameter.

Similar to the previous methods, but this method takes care of the high-level logic to directly construct a (possibly nonlinear) parameter from the input option.

Template Parameters

T	Parameter type. See Statically shaped tensors for supported types.

Parameters

name	Name of the model parameter.
input_option_name	Name of the input option that defines the value of the model parameter.
allow_nonlinear	Whether allows variable coupling

Returns

T The value of the registered model parameter.

declare_parameter() [2/3]

template<typename T , typename >

const T & declare_parameter ( const std::string & name, const T & rawval )

protected

Declare a parameter.

Note that all parameters are stored in the host (the object exposed to users). An object may be used multiple times in the host, and the same parameter may be declared multiple times. That is allowed, but only the first call to declare_parameter constructs the parameter value, and subsequent calls only returns a reference to the existing parameter.

Template Parameters

T	Buffer type. See Statically shaped tensors for supported types.

Parameters

name	Buffer name
rawval	Buffer value

Returns

Reference to buffer

declare_parameter() [3/3]

template<typename T , typename >

const T & declare_parameter ( const std::string & name, const TensorName< T > & tensorname, bool allow_nonlinear )

protected

Declare a parameter.

Similar to the previous method, but additionally handles the resolution of cross-referenced tensor name.

Template Parameters

T	Parameter type. See Statically shaped tensors for supported types.

Parameters

name	Name of the model parameter.
tensorname	The cross-ref'ed "string" that defines the value of the model parameter.
allow_nonlinear	Whether allows variable coupling

Returns

T The value of the registered model parameter.

get_parameter() [1/2]

TensorValueBase & get_parameter

(

const std::string &

name

)

Get a writable reference of a parameter.

get_parameter() [2/2]

const TensorValueBase & get_parameter ( const std::string & name ) const

inline

}@

Get a read-only reference of a parameter

named_parameters() [1/2]

std::map< std::string, std::unique_ptr< TensorValueBase > > & named_parameters

(

)

named_parameters() [2/2]

const std::map< std::string, std::unique_ptr< TensorValueBase > > & named_parameters ( ) const

inline

Returns

the buffer storage

operator=() [1/2]

ParameterStore & operator= ( const ParameterStore & )

delete

operator=() [2/2]

ParameterStore & operator= ( ParameterStore && )

delete

send_parameters_to()

void send_parameters_to ( const TensorOptions & options )

protectedvirtual

Send parameters to options.

Parameters

options

The target options

set_parameter()

void set_parameter	(	const std::string &	name,
		const Tensor &	value )

Set the value for a parameter.

set_parameters()

void set_parameters

(

const std::map< std::string, Tensor > &

param_values

)

Set values for parameters.